Description
The IDWT function is the inverse of DWT(td_dwt_mle). IDWT applies inverse wavelet transforms on multiple sequences simultaneously. IDWT takes as input the output table and meta table generated by DWT(td_dwt_mle) and outputs the sequences in time domain. Because the IDWT output is comparable to the DWT(td_dwt_mle) input, the inverse transformation is also called the reconstruction.
Usage
td_idwt_mle ( coefficient = NULL, meta.table = NULL, input.columns = NULL, sort.column = NULL, partition.columns = NULL, coefficient.sequence.column = NULL, meta.table.sequence.column = NULL )
Arguments
coefficient |
Required Argument. |
meta.table |
Required Argument. |
input.columns |
Required Argument. |
sort.column |
Required Argument. |
partition.columns |
Optional Argument. |
coefficient.sequence.column |
Optional Argument. |
meta.table.sequence.column |
Optional Argument. |
Value
Function returns an object of class "td_idwt_mle" which is a named list
containing Teradata tbl objects.
Named list members can be referenced directly with the "$" operator
using following names:
output.table
output
Examples
# Get the current context/connection con <- td_get_context()$connection # Load example data. # This example uses hourly climate data for five cities on a given day. loadExampleData("dwt_example", "ville_climatedata") # Create remote tibble objects. ville_climatedata <- tbl(con, "ville_climatedata") # Example 1 - # Apply DWT to sequences to create their coefficients and corresponding metadata. td_dwt_out <- td_dwt_mle(data = ville_climatedata, input.columns = c('temp_f','pressure_mbar','dewpoint_f'), sort.column = "period", wavelet = "db2", partition.columns = c("city"), level=2 ) # use the coefficient model table and the meta table generated by td_dwt_mle function # and apply td_idwt_mle to the filtered coefficients to reconstruct the sequences. td_idwt_out <- td_idwt_mle(coefficient = td_dwt_out$coefficient, meta.table = td_dwt_out$meta.table, input.columns = c("temp_f","pressure_mbar","dewpoint_f"), sort.column = "waveletid", partition.columns = c("city") )